Catálogo de publicaciones - libros

This paper describes a general framework for the detection and tracking of traffic and road signs from image sequences using only color information. The approach consists of two independent parts. In the first we use a set of Gaussian distributions that model each color for detecting road and traffic signs. In the second part we track the targets detected in the first step over time. Our approach is tested using image sequences with high clutter that contain targets with the presence of rotation and partial occlusion. Experimental results show that the proposed system detects on average 97% of the targets in the scene in near real-time with an average of 2 false detections per sequence.

This paper describes a new planning algorithm to perform automatic dimensional inspection of 3D industrial parts using a machine vision system. Our approach makes use of all the information available in the system: model of the machine part to inspect and characteristics of the inspection system obtained in previous calibration stages. The analysis is based on discretizing both the configuration space of the system as well as the geometry of the part. Our method does not limit the range of application of the system. It neither imposes any restrictions to the viewpoints from where the part is to be inspected nor to the complexity of the part. All the results shown in this study have been tested with a machine vision inspection system for quality control of industrial parts which consists of a stereoscopic pair of cameras and a laser plane.

Mixed Reality (MR) applications tend to focus on the accuracy of registration between the virtual and real objects of a scene, while paying relatively little attention to the representation of the luminance range in the merged video output. In this paper, we propose a means to partially address this deficiency by introducing Enhanced Dynamic Range Video, a technique based on differing brightness settings for each eye of a video see-through head mounted display (HMD). First, we construct a Video-Driven Time-Stamped Ball Cloud (VDTSBC), which serves as a guideline and a means to store temporal color information for stereo image registration. Second, with the assistance of the VDTSBC, we register each pair of stereo images, taking into account confounding issues of occlusion occurring within one eye but not the other. Finally, we apply luminance enhancement on the registered image pairs to generate an Enhanced Dynamic Range Video.

A major problem in surveillance systems is the occurrence of false alarms which lead people to take wrong actions. Thus, if the false alarm is frequent and occurs mainly due to system misclassification, this system will turn into an unreliable one and briefly out of use.

This paper proposes a classification method to oil spill detection using SAR images. The proposed methodology uses boosting method to minimize misclassification and also reach better generalization in order to reduce false alarms. Different feature sets were applied to single neural network classifiers and its performance were compared to a modified boosting method which provides a high-level feature selection. The experiments show substantial improvement in discriminating SAR images containing oil spots from the look-alike ones.

Paleography experts spend many hours transcribing ancient documents and state-of-the-art handwritten text recognition systems are not suitable for performing this task automatically. We propose here a new interactive, on-line framework which, rather than full automation, aims at assisting the experts in the proper recognition-transcription process; that is, facilitate and speed up the transcription of old documents. This framework combines the efficiency of automatic handwriting recognition systems with the accuracy of the experts, leading to a cost-effective perfect transcription of ancient manuscripts.

Degradation in old documents has been a matter of concern for a long time. With the easy access to information provided by technologies such as the Internet, new ways have arisen for consulting those documents without exposing them to yet more dangers of degradation. While restoration methods are present in the literature in relation to text documents and artworks, little attention has been given to the restoration of ancient music. This paper describes and compares different methods to restore images of ancient music documents degraded over time. Six different methods were tested, including global and adaptive thresholding, color clustering and edge detection. In this paper we conclude that those based on the Sauvola’s thresholding algorithm are the better suited for our proposed goal of ancient music restoration.

To a large extent noise suppression algorithms have been designed to deal with the two most classically defined types of noise: impulsive and Gaussian noise. However digitized images such as those acquired from historical photographs such as albumen prints contain a form of quasi-noise we shall term noise. This paper describes the concept of chaotic noise and proposes two fuzzy filters to suppress various types of noise in historical photographs based on the truncated median, an approximation of the mode filter.

This paper presents way of generation color documents from elements extracted from an original document image. This scheme yields synthetic documents of similar visual information to the original one. This method has several advantages as it allows a far more efficient way of storing and transmitting the information. Its rationale is to decompose a color document into paper and ink texture parameters and textual information. The textual information may be either typed or handwritten and is stored as a compressed monochromatic image.

Portable digital cameras are of widespread use today. Their image quality, low cost and portability have drastically changed the culture of photography. Professionals of many different areas start to take photos of documents, instead of photocopying them. This article presents techniques for improving the quality of document images digitalized with portable digital cameras.

In this paper, we propose a coordinate transform model to rectify both the perspective and surface curving distortions of a book image, which is captured by a 2D digital camera. The rectification process includes three steps: 1) detect the boundary of a curved page on a book image; 2) knowing the four boundary curves of a book page should be mapped to the four sides of a rectangle, we build up the correspondence between a pixel inside the book page and a pixel inside the rectangle; 3) fill each pixel in the rectangle with its correspondence pixel on the book image, we get a rectangle which is the dewarped page we want. The experimental results show the effectiveness of the proposed rectification approach.